A Two-Level Distributed Architecture for Efficient Web Content Adaptation and Delivery

  • Authors:
  • Claudia Canali;Valeria Cardellini;Michele Colajanni;Riccardo Lancellotti;Philip S. Yu

  • Affiliations:
  • University of Parma;University of Roma "Tor Vergata";University of Modena and Reggio Emilia;University of Modena and Reggio Emilia;IBM T.J. Watson Research Center

  • Venue:
  • SAINT '05 Proceedings of the The 2005 Symposium on Applications and the Internet
  • Year:
  • 2005

Quantified Score

Hi-index 0.00

Visualization

Abstract

The complexity of services provided through the Web is continuously increasing as well as the variety of new devices that are gaining access to the Internet. Tailoring Web and multimedia resources to meet the user and client requirements opens two main novel issues in the research area of content delivery. The working set tends to increase sub-stantially because multiple versions may be generated from the same original resource. Moreover, the content adaptation operations may be computationally expensive. In this paper, we consider third-party infrastructures composed by a geographically distributed system of intermediary and cooperative nodes that provide fast content adaptation and delivery of Web resources. We propose a novel distributed architecture of intermediary nodes which are organized in two levels. The front-end nodes in the first tier are thin edge servers that locate the resources and forward the client requests to the nodes in the second tier. These interior nodes are fat servers that run the most expensive functions such as content adaptation, resource caching and fetching. Through real prototypes we compare the performance of the proposed two-level architecture to that of alternative one-level infrastructures where all nodes are fat peers providing the entire set of functions.